Show simple item record

dc.contributor.authorSo, Stephenen_US
dc.contributor.authorPaliwal, Kuldipen_US
dc.contributor.editorLuís Oliveiraen_US
dc.date.accessioned2017-05-03T13:01:32Z
dc.date.available2017-05-03T13:01:32Z
dc.date.issued2005en_US
dc.date.modified2007-03-21T21:26:49Z
dc.identifier.refurihttp://www.interspeech2005.org/en_AU
dc.identifier.urihttp://hdl.handle.net/10072/2673
dc.description.abstractIn this paper, we examine a coding scheme for quantising feature vectors in a distributed speech recognition environment that is more robust to noise. It consists of a vector quantiser that operates on the logarithmic filterbank energies (LFBEs). Through the use of a perceptually-weighted Euclidean distance measure, which emphasises the LFBEs that represent the spectral peaks, the vector quantiser codebook provides /emph{a priori} knowledge of the spectral characteristics of clean speech and is used to quantise features from noise-corrupted speech. Our comparative results from the ETSI Aurora-2 recognition task show that the perceptually-weighted vector quantisation of LFBEs achieves higher recognition accuracies for noisy speech than the unweighted vector quantisation, memoryless and multi-frame GMM-based block quantisation and scalar quantisation of Mel frequency-warped cepstral coefficients.en_US
dc.description.peerreviewedYesen_US
dc.description.publicationstatusYesen_AU
dc.languageEnglishen_US
dc.language.isoen_AU
dc.publisherInternational Speech Communication Association (ISCA)en_US
dc.publisher.placeLisbon, Portugalen_US
dc.relation.ispartofstudentpublicationNen_AU
dc.relation.ispartofconferencename9th European Conference on Speech Communication and Technologyen_US
dc.relation.ispartofconferencetitleInterspeech 2005 - Eurospeechen_US
dc.relation.ispartofdatefrom2005-09-04en_US
dc.relation.ispartofdateto2005-09-08en_US
dc.relation.ispartoflocationLisbon, Portugalen_US
dc.rights.retentionYen_AU
dc.subject.fieldofresearchcode280204en_US
dc.subject.fieldofresearchcode280206en_US
dc.titleImproved noise-robustness in distributed speech recognition via perceptually-weighted vector quantisation of filterbank energiesen_US
dc.typeConference outputen_US
dc.type.descriptionE1 - Conference Publications (HERDC)en_US
dc.type.codeE - Conference Publicationsen_US
gro.facultyGriffith Sciences, Griffith School of Engineeringen_US
gro.date.issued2005
gro.hasfulltextNo Full Text


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

  • Conference outputs
    Contains papers delivered by Griffith authors at national and international conferences.

Show simple item record